Tradition retinal detachments secondary to periretinal and intravitreal cellular membranes are the major cause of failure following retinal detachment surgery and penetrating trauma. Clearly and Ryan developed a model for posterior penetrating trauma, demonstrating the importance of serum in the pathogensis of traction retinal detachment. We have previously shown that serum contains chemoattractants for a number of cells including retinal pigment epithelial cells, fibroblasts and astrocytes, and that fibronectin (FN) and platelet-derived growth factor (PDGF) are two of the active components in serum effecting this migration. Surgical manipulation, cryotherapy & trauma have been shown to produce a breakdown of blood-ocular barrier with the subsequent presence of serum in the vitreous cavity. PDGF and FN within the eye attract cells into the vitreous where proliferation, and membrane formation & contraction then cause traction detachments. To support this theory we developed a rabbit model where intravitreally injected PDGF and FN in conjunction with a penetrating wound reproducibly resulted in the development of traction retinal detachment. We obtained retinal detachments within four weeks in 13 of 18 experimental eyes as compared with 3 of 16 controls. (P greater than 0.0025 by Fisher's exact test). In this study we will develop an experimental model for traction retinal detachment in the sub-human primate. In order to induce the formation of cellular membranes in a manner similar to the human disease we will use PDGF and FN as defined biochemical mediators. We will create a standard scleral laceration that will consist of a 8mm parsplana incision with vitreous prolapse with two subsequent intravitreal injections of PDGF and FN. Along with the establishment of the model we will study the optimal doses of PDGF and FN necessary to stimulate the formation of intraocular membranes and traction retinal detachment. We will also determine the importance of PDGF and FN in the pathogenesis of PVR following giant retinal tears. For this purpose giant retinal tears will be created and PDGF and FN injected intravitreally. In all cases, the animals will be closely monitored by indirect ophthalmoscopy and fundus photography. After the proposed 12 week observation period eyes will be enucleated and the histopathology of the wound, vitreous and retina studied at the light microscopic level. The development of such a straightforward and reliable model with close resemblance to the human disease process will provide a valuable tool to advance our understanding of the pathogenesis of PVR and for the uniform evaluation of new therapeutic modalities.